[1979] in Kerberos_V5_Development
Re: -Wall flamage
daemon@ATHENA.MIT.EDU (Barry Jaspan)
Tue Nov 19 17:30:38 1996
Date: Tue, 19 Nov 1996 17:30:16 -0500
From: "Barry Jaspan" <bjaspan@MIT.EDU>
To: tlyu@MIT.EDU
Cc: krbdev@MIT.EDU
In-Reply-To: <9611192210.AA27291@tesla-coil.MIT.EDU> (message from Tom Yu on
Tue, 19 Nov 1996 17:10:40 -0500)
Date: Tue, 19 Nov 1996 17:10:40 -0500
From: Tom Yu <tlyu@MIT.EDU>
We should try to eliminate at least *some* of the warnings issued by
gcc -Wall, among them things like "control reaches end of non-void
function" and "return with no value in function returning non-void",
etc.
If gcc -Wall discovers bugs that have real consequences, sure, they
should be fixed. But unless it is totally obvious that it needs to be
fixed and how to fix it, perhaps you should submit a PR so the bug can
go through the group priority-assignment process. Or, maybe with
three business days until code freeze, we don't have time for
that... I'd say just avoid the temptation to fix things that do not
need to be fixed since you may break something worse.
Barry
